Good Games, Bad Design - Episode 1: What's at Stake

By Eric-Jon Rössel Tairne [07.09.10]

[In the first part of a new GameCareerGuide series, Eric-Jon Rössel Tairne analyzes a good game that doesn't quite hit the mark in terms of its overall design -- starting with the NES classic Castlevania III.]

There are two basic ways that video games communicate ideas -- through the actions allowed the player, and through the environment on which the player may act. The player's every action changes the player's immediate relationship with the environment, which in turn shapes the player's potential for action. Let's say you shoot an asteroid. Although the immediate obstacle is gone, now you've several smaller rocks to deal with, moving faster, in different trajectories.

The more you do, and the more feedback the game gives you, the more you adapt your behavior. When an action results in success or a reward, you tend to repeat it. When you get an unpleasant result, you tend to avoid repeating yourself.

A successful game environment does four things:

it teaches about the player's relationship with the environment;
in doing so, it directs and focuses the player's behavior;
generally it obscures this manipulation from the player; and so
through the invoked behavior it evokes in the player a certain mood or mindset.

If the player doesn't know why he picks the routes and actions he does, yet in picking those routes and actions he comes to adopt the intended perspective, you have successfully communicated. Think of all the moments in Half-Life 2 where you think you're being clever under pressure, and you're actually choosing the only possible path -- or how The Legend of Zelda keeps you on-track by making the woods scary and dangerous, so that you will tend to leave them until you're stronger and more experienced.

Is level design everything? Only if your game has something to say. If you're retreading old ground, and you expect the audience knows the routine, then you can toss them any old nonsense. Of course then few of the player's actions will have real consequence, so the game will feel unresponsive and dull. Still, maybe if you add some flashy features or cutscenes you can distract the player for a while. If you're afraid of putting people off, you can patronize them with elaborate tutorials.

There's no fooling the outsiders, though. If your game fails to communicate on its own merits, then no one besides the fans will bother with it. And even within that audience the conversation will narrow and turn from big, nourishing ideas to minutiae -- as if the differences between one leveling system and the next really matter in themselves. This heads-down view leads us away from meaningful representation, and toward thoughtless copying and repetition, abstracted and regimented genres, fractured markets, and eventually a whole medium that is impenetrable to outside eyes.

As in any human endeavor, sloppy or thoughtless design is perhaps more the rule than the exception. And that's fair enough, when that design is a part of a lousy game that no one is likely to take seriously. More worrisome are the otherwise good, solid games that a student of design may well look to for inspiration. Games don't have much of a critical history; their culture treats anything "good" as model of perfection that everything new should strive to imitate down to the pixel. It's hard to break out of that mindset, and to look at design in terms of problems and solutions.

A solution, of course, only makes sense in context. In a game, each mechanism serves to illustrate to the player some concept, or to solve a logistical problem in the game's premise. Anything that serves neither of these purposes is extraneous -- and the key to communication is if you don't need it, cut it out. It is in this spirit that some case models may be illustrative.

Dracula's Curse

The first NES Castlevania is groundbreaking, rhythmic, and intricately designed. The second is a radical departure, expanded from the first game's themes but structured unlike anything before or since. Castlevania III is exactly like the first, except enormous. The game's purpose is to take up as much space as possible, so as to give the player a sense of grandeur. The trade-off is that from moment to moment there isn't much attention to detail or flow, leading the player to waste time a) maneuvering between points of interest, b) exploring false leads, and c) replaying poorly balanced sequences.

Unlike either of its predecessors, there is little reward for being observant and exploring, and the game lacks those little reward beats and that clear direction that builds momentum. The architecture is full of empty spaces, meaningless flourishes, and padding. Creatures are tossed around with little thought as to placement. The branching paths and multiple characters are new, but compared to the focused, driving architecture of the original Castlevania, there is little psychology to the design. So those ideas rather go to waste, as the game never clearly builds on them.

Click image for full size.

Let's break down the first level. Over about 21 screens, it meanders through a courtyard, up a ruined abbey, along a wall, down past a ghost town, and into a cemetery. The level divides into four blocks, each separated by a door; each block might consist of one to four horizontally or vertically scrolling areas, each area about three screens in size. We begin as our protagonist Trevor rises from prayer at an ornate shrine.

Click image for full size.

LEVEL 1-01: You walk to the right; what does staircase structure signify? There is no compulsion or benefit to climb it. It does little to pace the level. The architecture is functionally dead -- yet appears to be the only feature of this room. When you've passed it, the room is over. What, if anything, have we learned of the game's controls, themes, premise, pacing, or atmosphere?

Instead: remember the staircase in level one of the original Castlevania? It was a branching choice: stay on the ground with the endless zombies, or climb up and take your chances with less predictable enemies. Maybe earn a reward in the process. Considering that Castlevania III is supposed to be all about branching paths, here's a missed opportunity to establish that right out of the gate.

LEVEL 1-02: So having learned nothing, you pass to the second room to find two lethargic skeletons and a double staircase, with a small platform at the kink. At the top of that staircase is a larger platform, with another staircase exiting the top of the screen.

Presumably this screen serves to teach the player about stairs, in case the previous room was too vague. You can't get anywhere by jumping, so you need to find another route. Yet the actual arrangement is careless.

The screen is mostly empty, with a double-thick wall of blocks to the right. The wall is far enough from the "action" as to feel odd. The emptiness of the composition draws attention away from the stairs, and toward another meaningless clump of bricks. In either of the game's predecessors, this would be a clue to a secret reward. Here, exploration of the bricks is fruitless.

This sort of misleading dead space is everywhere, so for shorthand let's call them "disappointment wells." While we're here the upper platform holds another, slightly less distracting, such well.

Consider this: instead, have two entrances to the room. One door is high, the other low. Perhaps they could branch from that stairwell in 1-01. Ideally from here the level itself could further branch. Players could either climb the clock tower, or head off to the right, perhaps through a cave or up a grassy hill. Toward the end of the level, the paths would merge. Then after beating the boss, the game would (as now) present a choice of entire levels. The difference is that the player would have built up to that decision.

Or, let's say we keep the level structure intact. Instead, each route through this room would have its own sacrifice and payoff. If you come in through the bottom door, maybe there's an extra enemy or two. Maybe that provides the player with flashy bonus or (better) an axe, to make the next section easier. Entering through the top door would lead directly to the chapel, but would miss out on all the interesting stuff.

The chapel is mostly fine. Toward the top there's that odd platform to the left, that lures the player to jump down, again to no purpose. The platform might be salvaged if there were some associated bonus. The player can choose between going out of his way for a reward, or just plodding forward, demonstrating again on a small scale the game's broader cost/benefit structure.

At the top of the stairs, a skeleton flies into our face. Why does the first enemy to pose any real threat leap directly at the player's position, and why is it so difficult to dodge? If forewarned, you can avoid it; the first couple of times, it's almost a mandatory hit. What does this enemy placement teach us? Well, nothing. It's just obnoxious.

We also see two arbitrary floors, with a disappointment well to the right. Instead of leading up and right, to maintain the player's momentum, the stairs lead left. Why? If it's meant to continue the spiraling motion of the chapel, why force the player to so far to the right to ascend?

The next two skeletons (not pictured) are placed well enough, though the one on the upper level might be better swapped for something with a more direct attack pattern. Down and to the left we've another small disappointment well. Then over the sensible gap in the floor is one of the game's more inexplicable moments: an inaccessible two-block platform, with a candle on top.

There is no way to both destroy this candle and claim its contents. Why the platform? If it's just decoration, okay; it's clear the player can't jump to it, so fair enough. So then why the candle? This is just baffling. It makes me think that the designers intended that the player could return here with the wall-climbing pirate Grant -- a feature not present in the final game.

Now, if there were no platform and if the player happened to hang onto an axe from that first (hypothetical) split path, perhaps he could flick it here to gain a special reward for his foresight -- maybe a guaranteed hit multiplier.

Click image for full size.

Down the stairs, to a pair of disappointment wells to the left and another to the upper-right. More baffling empty space. The first swivel platform is well-placed, acting in a positive role as a shortcut before the more sinister placement to come.

I'll point out here that this block has been dragging on for ages, and if a Medusa head knocks you down a pit here you have to climb all the way back up the chapel again. And for what reward? Why, yet another disappointment well.

Granted, the lower level also serves as a safety net for the upper level, allowing you to adapt to the swivel platforms. Yet the layout makes this unclear; both levels are straight, and the bottom has a higher ceiling. The incidental placement of the stairs makes the paths look equally valid. If you want a safety net, perhaps wall off the bottom path but make the wall easy to scale from the inside.

Click image for full size.

LEVEL 1-03: Finally another checkpoint, and a long straightaway. This area mostly serves to show off some neat background tiles.There are a couple of random pits that, due to the ground's inconsistent tiling pattern and the unexpected lack of a wall, aren't clearly flagged as pits. There might be something neat down there! There's a random elevated platform that you can't climb onto and that serves no other purpose. This is a nice, relaxing beat. A shame it doesn't say anything useful. We're getting close to the end, and the level has yet to actively engage the player. In place of zombies, this would be a good place to practice with aerial enemies, if there were any.

Click image for full size.

Down the stairs we have an instant disappointment well, followed by the one true hidden reward in this level, a pork chop under the stairs. After all that earlier misdirection, and the extra steps in claiming the prize, it feels a little underwhelming. Not only do you have to whip through two barriers; at the same time you also have to field a fleaman.

Incidentally, we have already seen random jumping skeletons -- so why fleamen? They're the same thing, except more annoying. What do they serve to teach us at this tender stage of indoctrination? Imagine if Lakitu were in the first level of Super Mario Bros. Same question with the long stretch of ground punctuated with a random Castlevania-brand dragon-head-cannon thing. Maybe it serves to show that some enemies take more than one hit. Probably not.

If the paths had split back in level 1-02, they might merge here. Since this is a town and we've paths to avoid the ground level, this would be a better place for those endless zombies from the previous area. The rhythm of zombie whipping would serve as a nice physical ramp-up for the boss.

This segment also shows a weird inconsistency to the game's collision. Sometimes when you jump up, as on the first stair structure to the right, Trevor's head will pass through the blocks, allowing you to jump normally. Sometimes, as with the following lower structure, he will hit his head or he can't jump at all. Why? The game is never completely clear.

Click image for full size.

LEVEL 1-04: A long, flat run-up to the boss. Nothing wrong here; it's a convention for the level architecture to grow long and quiet before a boss, to let you know that something big is coming up. As conventions go, it's a solid one -- a simple trick of pacing and psychology, that serves its purpose.

As for the boss area, though, it is very difficult to avoid being hit at least once. You have to walk past the middle of the screen before the boss even awakens. If you keep walking, you will be trapped at the right side of the screen and suffer many blows. If you double back, with a mind to playing follow-the-leader around the raised platform to the left, the boss will probably hit you at least once as you pass him. Not good; especially for the first major battle in the game, the player should have at least a chance of coming through unscathed. Either move the "safe area to" the right, or build in a passage to allow the player to scurry back to the more advantageous position at left.

And there we have it.

Blood from a Stone

In a good design, you should be able to take any one beat and break it down into the game's basic themes. The minutiae reinforce the overall structure, and the overall structure lends purpose to the minutiae.

What annoys me personally about Dracula's Curse is that after the abrupt change of format in Simon's Quest, this game reverts hard to the shape of the original game and ladles on the raw content -- yet it handles the content so ineptly that it makes play feel like work. I can excuse a conservative approach as an exercise of craft over art -- yet that laborious attention to detail is absent. Pound for pound, level for level, Castlevania III is a poorer game than the original, and there's far more of it to digest.

Which isn't to say that it's a bad game at all. The controls are as crunchy as any Konami game from that era. The music is splendid (if again broader than it is deep). The choose-your-own-adventure structure and character switching are neat gimmicks on their own. It's just a shame that it's all so uninterested in conversation.

The best bits never amount to more than gimmicks, and from moment to moment the game rarely acknowledges, much less rewards, the player's efforts. You play because you're the player, and that's what a player does. The game shrugs and provides some things to play through. That's the sum of the relationship.

There's a place for empty relationships, but they'll never make a list of the top romances. If you're starting from scratch, you might as well pursue something nourishing and make the world a little bit richer.